Personalized Entity Repository Segmentation for Mobile Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile devices face challenges in providing personalized user experiences due to limited storage and the need for constant connectivity to large public entity repositories, which hinders efficient entity recognition and assistance capabilities.
Innovation Solution
Dividing the entity repository into fixed, location-based, topic-based, or functional sets, allowing a client device to determine and download only relevant sets, using a set identification engine with a prediction model to rank and manage these sets for optimal storage and usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a mobile device uses a large public entity repository to provide personalized user experiences, then the user experience personalization is improved, but the device storage requirement increases and constant network connectivity is needed
Solution Approach 1:
The entity repository is divided into multiple fixed sets or slices, where each set contains entities related to specific topics, locations, or functions. The mobile device downloads only the relevant sets needed for personalized experiences rather than the entire repository, thereby reducing storage requirements while maintaining personalization capabilities.
Solution Approach 2:
The system extracts and downloads only the necessary portions of the entity repository (relevant fixed sets) to the mobile device based on user context, location, and usage patterns. This extraction approach allows the device to have access to personalized entity information without storing the complete repository.
2Reliability
If a mobile device downloads complete entity repository to ensure offline access, then offline entity recognition is improved, but the device storage and data transfer increase
Solution Approach 1:
The entity repository is segmented into fixed sets that can be selectively downloaded. This allows the device to obtain offline access to specific entity collections relevant to user needs without transferring the entire repository, thus reducing data transfer while ensuring offline recognition capability.
Solution Approach 2:
The system performs preliminary determination of which fixed sets are relevant to the user based on context analysis, and downloads only those sets in advance. This preliminary selection action ensures that the device has the necessary entity data for offline operation without unnecessary data transfer.
3Measurement precision
If a mobile device stores multiple fixed sets of entities, then the entity recognition accuracy is improved, but the storage space consumption increases
Solution Approach 1:
Different fixed sets of entities are stored based on local relevance to the user's context, location, and usage patterns. The device maintains high entity recognition accuracy for relevant domains by storing specialized entity sets for those areas, while avoiding storage of entities for irrelevant domains, thus optimizing storage space utilization.
Solution Approach 2:
The device stores a partial collection of entity sets that are most relevant to current user needs rather than attempting to store all possible entity sets. This partial action approach maintains sufficient recognition accuracy for the user's context while constraining storage space consumption.
4Loss of information
If a mobile device frequently updates entity repository to maintain current information, then the information freshness is improved, but the network connectivity requirement and energy consumption increase
Solution Approach 1:
The system performs periodic updates of fixed sets based on their relevance and the need for information freshness. Rather than continuous synchronization, updates are triggered periodically or based on specific conditions, reducing energy consumption while maintaining information currency for relevant entity sets.
Solution Approach 2:
The system uses feedback from usage patterns, location changes, and context analysis to determine which fixed sets need updating. This feedback-driven update mechanism ensures that energy is spent only on refreshing entity sets that are currently relevant to the user, rather than uniformly updating all sets.
Data Source
AI summary
Systems and methods are provided for a personalized entity repository. For example, a computing device comprises a personalized entity repository having fixed sets of entities from an entity repository stored at a server, a processor, and memory storing instructions that cause the computing device to identify fixed sets of entities that are relevant to a user based on context associated with the computing device, rank the fixed sets by relevancy, and update the personalized entity repository using selected sets determined based on the rank and on set usage parameters applicable to the user. In another example, a method includes generating fixed sets of entities from an entity repository, including location-based sets and topic-based sets, and providing a subset of the fixed sets to a client, the client requesting the subset based on the client's location and on items identified in content generated for display on the client.